STQA - Human-Authored Questions - Forecast MRE: leaderboard
Metric: Mean Relative Error. Source: arxiv.org. Saturation forecast: Around March 2027. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-OSS-120B | 0.07 |
| 2 | GPT-OSS-20B | 0.07 |
| 3 | Qwen 3 30B A3B | 0.07 |
| 4 | Gemma 4 31B | 0.08 |
Interactive version: theaggregate.ai/benchmark?slug=stqa-human-authored-questions-forecast-mre · How It Works · Data refreshed daily, snapshot 2026-09-24.